Data-Driven Behavioural Analytics in Web Logs Using Machine Learning Techniques

Abstract

Behaviour analysis of users on the internet is a crucial area that enables various features can be studied via user behaviour on the internet. The intention prediction is been recent research that identifies the user interactions on a website. Additionally, addressing the demand and enabling the information prediction for users enforces analysis of the user navigation behaviour. In this paper, we study the user behaviour in e-marketing websites to increase the relevance of bringing the products based on the user behaviour. The study uses a machine-learning algorithm with several metrics that studies the logs of several users during the training phase and provides user-specific relevant information during the testing phase. The simulation is conducted to test the efficacy of machine learning in providing the results relevant to the user behaviour, where accuracy is the main metric that is tested to address the machine learning ability. From the results, it is found that the proposed machine learning model achieves a higher rate of accuracy than other existing methods